High-dimensional bootstrap and asymptotic expansion

The recent seminal work of Chernozhukov, Chetverikov and Kato has shown that bootstrap approximation for the maximum of a sum of independent random vectors is justified even when the dimension is much larger than the sample size. In this context, numerical experiments suggest that third-moment match bootstrap approximations would outperform normal approximation even without studentization, but the existing theoretical results cannot explain this phenomenon. In this paper, we first show that Edgeworth expansion, if justified, can give an explanation for this phenomenon. Second, we obtain valid Edgeworth expansions in the high-dimensional setting when the random vectors have Stein kernels. Finally, we prove the second-order accuracy of a double wild bootstrap method in this setting. As a byproduct, we find an interesting blessing of dimensionality phenomenon: The single third-moment match wild bootstrap is already second-order accurate in high-dimensions if the covariance matrix is spherical.